aquisition of parts of large signals with ni scope

[TimT2006]

Hello

I am trying to aquire large data with ni scope.

The aquisition time is about 2 seconds with a sampling frequency of 50 MHz.

I figured out that this amount of data is too large. The NI USB 5133 does not have enough memory.

Actually I do only need some parts of this 2senconds signal thus it would be possible to aquire only some parts of the signal.

Moreover I need to average the signal to reduce the noise.

I am trying and trying but couldn't make it out so far.

Maybe you could help me with some ideas....

Thank you very much

[TimT2006]

I found the option that I can use the property node "fetch offset"....

but there is still an error message saying there is an overload in the memory.

[TimT2006]

If i am using the option "fetch offset" I recognized that it is very slow.

Do you have any ideas for me to split the complete signal into parts to aquire to hole signal?

Thank you...

[Ken_Tobler]

Hello

The Scopes are limited by the amount of internal memory where they can stock the samples.
It means in your case you can try to generate a trigger which start the acquisition only in the important parts of you signal or you can try with this example:
http://zone.ni.com/devzone/cda/epd/p/id/5427

It gives the possibility to acquire more samples than the internal buffer. It means that you can configure more records as the memory allows (create a circular buffer) but then you should be enough fast to read the records out of the memory.

Best regards
Ken

[TimT2006]

HI there...

Thank you for you answer.

Unfortunately I do not have Labview 8.5 - Thus I cannot open this file: multi-records_fetch.vi

But somehow there must be a possibility to save a aignal that is larger than 4Mb?

I tried to solve the problem using the trigger delay.

I tried to shift the delay and add all the parts later on.... but didn't work.

I added my original Vi (Labview version 8.2.1) to my message.

[TimT2006]

Hi...

I generated a trigger which always occurs just before all the significant parts of my signal.

My problem now: How can I recognize this and how can I just record those parts after the trigger?

I can't open this file: http://zone.ni.com/devzone/cda/epd/p/id/5427 - it is written with LV 8.5 - I only have 8.2

Thank you

[Murici]

Hi TimT20006,

I have the same problem as yours. The NI USB 5133 has 4Mb onboard memory and I need to acquire at least 1 second time window with about 50MS/s with 2 separeted channels. As we faced some difficulties with memory, we stated looking here in the forum and found your post. ? Was your problem already solved with the  multi-records_fetch.vi ? perhaps you found another solution....

Thanks,

Guilherme Murici

[Jennifer O]

This type of question would be more visible on the High Speed Digitizers Forum.  

The USB-5132/5133 is a finite acquisition digitizer.  The record size is limited by the total amount of memory on the digitzer.  A 32MB/ch memory option for the USB-5132 and USB-5133 has just been released, which increases the maximum record length by a factor of 8 in comparison to the 4MB/ch option. 

If you need to acquire more data, you must re-initiate the task to take a second record.  There will be some "deadtime" where no samples are acquired during this initiate step.  Some benchmarks have been provided to show what equivalent data rate can be achieved using this method.  For continuous acquisition, consider a PCI, PXI or PXIe digitizer.

Jennifer O.